# Mathematical Reasoning on AIME25 decontaminated (Accuracy)

Source: [Mathematical Reasoning on AIME25 decontaminated (Accuracy)](<https://www.sota2.com/research/sota/mathematical-reasoning-on-aime25-decontaminated-accuracy>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-05-27T09:51:01.147Z

Evaluation of model performance on the decontaminated version of the AIME 2025 dataset for Mathematical Reasoning, reporting Accuracy.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Mathematical Reasoning |
| Dataset (including any stated split) | AIME25 decontaminated |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 14 of 14 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Accuracy](<https://www.sota2.com/research/sota/mathematical-reasoning-on-aime25-decontaminated-accuracy.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| DAPO++ | 22.6 | Not recorded | 6a167ac8dbd7150506f801ab |
| DAPO-Math-17k | 21.5 | Not recorded | 6a167ac8dbd7150506f801ab |
| DeepScaleR | 17.1 | Not recorded | 6a167ac8dbd7150506f801ab |
| DeepMath-103K | 16.7 | Not recorded | 6a167ac8dbd7150506f801ab |
| Skywork-OR1-RL-Data | 15.8 | Not recorded | 6a167ac8dbd7150506f801ab |
| OpenR1-Math-220k | 15.4 | Not recorded | 6a167ac8dbd7150506f801ab |
| Qwen3-8B-Base | 10.5 | Not recorded | 6a167ac8dbd7150506f801ab |
| DeepScaleR | 7.6 | Not recorded | 6a167ac8dbd7150506f801ab |
| OpenR1-Math-220k | 6.3 | Not recorded | 6a167ac8dbd7150506f801ab |
| DAPO-Math-17k | 5.9 | Not recorded | 6a167ac8dbd7150506f801ab |
| Skywork-OR1-RL-Data | 4.6 | Not recorded | 6a167ac8dbd7150506f801ab |
| DAPO++ | 3.3 | Not recorded | 6a167ac8dbd7150506f801ab |
| DeepMath-103K | 3.1 | Not recorded | 6a167ac8dbd7150506f801ab |
| Qwen3-1.7B-Base | 2.3 | Not recorded | 6a167ac8dbd7150506f801ab |

### DAPO++

Accuracy: 22.6. Result ID: 6a16bdaefbc373592c1051c3.

[Reporting paper](<https://www.sota2.com/research/paper/6a167ac8dbd7150506f801ab.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a167ac8dbd7150506f801ab.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-8B","Sampling Strategy":"Mean@32"}. This record does not assert independent reproduction.

### DAPO-Math-17k

Accuracy: 21.5. Result ID: 6a16bdaefbc373592c1051c1.

[Reporting paper](<https://www.sota2.com/research/paper/6a167ac8dbd7150506f801ab.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78a81e209131cea79ac6.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-8B","Sampling Strategy":"Mean@32"}. This record does not assert independent reproduction.

### DeepScaleR

Accuracy: 17.1. Result ID: 6a16bdaefbc373592c1051be.

[Reporting paper](<https://www.sota2.com/research/paper/6a167ac8dbd7150506f801ab.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-8B","Sampling Strategy":"Mean@32"}. This record does not assert independent reproduction.

### DeepMath-103K

Accuracy: 16.7. Result ID: 6a16bdaefbc373592c1051bf.

[Reporting paper](<https://www.sota2.com/research/paper/6a167ac8dbd7150506f801ab.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78b51e209131cea79e64.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-8B","Sampling Strategy":"Mean@32"}. This record does not assert independent reproduction.

### Skywork-OR1-RL-Data

Accuracy: 15.8. Result ID: 6a16bdaefbc373592c1051c2.

[Reporting paper](<https://www.sota2.com/research/paper/6a167ac8dbd7150506f801ab.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78eb1e209131cea7b377.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-8B","Sampling Strategy":"Mean@32"}. This record does not assert independent reproduction.

### OpenR1-Math-220k

Accuracy: 15.4. Result ID: 6a16bdaefbc373592c1051c0.

[Reporting paper](<https://www.sota2.com/research/paper/6a167ac8dbd7150506f801ab.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-8B","Sampling Strategy":"Mean@32"}. This record does not assert independent reproduction.

### Qwen3-8B-Base

Accuracy: 10.5. Result ID: 6a16bdaefbc373592c1051bd.

[Reporting paper](<https://www.sota2.com/research/paper/6a167ac8dbd7150506f801ab.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-8B","Sampling Strategy":"Mean@32"}. This record does not assert independent reproduction.

### DeepScaleR

Accuracy: 7.6. Result ID: 6a16bdaefbc373592c1051b7.

[Reporting paper](<https://www.sota2.com/research/paper/6a167ac8dbd7150506f801ab.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-1.7B","Sampling Strategy":"Mean@32"}. This record does not assert independent reproduction.

### OpenR1-Math-220k

Accuracy: 6.3. Result ID: 6a16bdaefbc373592c1051b9.

[Reporting paper](<https://www.sota2.com/research/paper/6a167ac8dbd7150506f801ab.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-1.7B","Sampling Strategy":"Mean@32"}. This record does not assert independent reproduction.

### DAPO-Math-17k

Accuracy: 5.9. Result ID: 6a16bdaefbc373592c1051ba.

[Reporting paper](<https://www.sota2.com/research/paper/6a167ac8dbd7150506f801ab.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78a81e209131cea79ac6.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-1.7B","Sampling Strategy":"Mean@32"}. This record does not assert independent reproduction.

### Skywork-OR1-RL-Data

Accuracy: 4.6. Result ID: 6a16bdaefbc373592c1051bb.

[Reporting paper](<https://www.sota2.com/research/paper/6a167ac8dbd7150506f801ab.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78eb1e209131cea7b377.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-1.7B","Sampling Strategy":"Mean@32"}. This record does not assert independent reproduction.

### DAPO++

Accuracy: 3.3. Result ID: 6a16bdaefbc373592c1051bc.

[Reporting paper](<https://www.sota2.com/research/paper/6a167ac8dbd7150506f801ab.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a167ac8dbd7150506f801ab.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-1.7B","Sampling Strategy":"Mean@32"}. This record does not assert independent reproduction.

### DeepMath-103K

Accuracy: 3.1. Result ID: 6a16bdaefbc373592c1051b8.

[Reporting paper](<https://www.sota2.com/research/paper/6a167ac8dbd7150506f801ab.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78b51e209131cea79e64.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-1.7B","Sampling Strategy":"Mean@32"}. This record does not assert independent reproduction.

### Qwen3-1.7B-Base

Accuracy: 2.3. Result ID: 6a16bdaefbc373592c1051b6.

[Reporting paper](<https://www.sota2.com/research/paper/6a167ac8dbd7150506f801ab.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-1.7B","Sampling Strategy":"Mean@32"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
